CAJ | 학술논문

为了减小对车辆进行外形超限测量时其他车辆对图像的影响，精确地将目标车辆的轮廓信息分割并测量，提出一种基于形状角、结合分块图像与边缘提取的分割算法。对获取图像进行帧差法削弱背景边缘信息的影响，再用Canny算子边缘提取，将得到的轮廓信息进行基于形状角的识别与提取。提出了一种对图像进行分块分类的图像分割法。该算法有效解决了测量中背景对前景的影响。实验结果表明该算法具有较高的鲁棒性与准确性。
위료감소대차량진행외형초한측량시기타차량대도상적영향，정학지장목표차량적륜곽신식분할병측량，제출일충기우형상각、결합분괴도상여변연제취적분할산법。대획취도상진행정차법삭약배경변연신식적영향，재용Canny산자변연제취，장득도적륜곽신식진행기우형상각적식별여제취。제출료일충대도상진행분괴분류적도상분할법。해산법유효해결료측량중배경대전경적영향。실험결과표명해산법구유교고적로봉성여준학성。
In order to reduce the influence of other vehicles on image,and segment and measure the outline information of the target vehicle accurately while the vehicle shape ultralimit measurement is executed,a shape angle based segmentation algo-rithm which is combined with block image and edge extraction is put forward in this paper. The frame differential method is adopted to weaken the influence of image edge information of the background,and then Canny operator edge detection is used to recognize and extract the obtained contour information on the basis of shape angle. An image segmentation method which classi-fies an image by segmenting the image into blocks is proposed. The algorithm effectively solves the influence of background on foreground during the measurement. Experimental results show that the algorithm has strong robustness and high accuracy.